IRPact_modellierung.distributions

Class FiniteMassPointsDiscreteDistribution

java.lang.Object

IRPact_modellierung.distributions.Distribution

IRPact_modellierung.distributions.UnivariateDistribution

IRPact_modellierung.distributions.BoundedUnivariateDistribution

IRPact_modellierung.distributions.FiniteMassPointsDiscreteDistribution

public class FiniteMassPointsDiscreteDistribution
extends BoundedUnivariateDistribution

Class that models a distribution that has its mass concentrated in a finite number of points. Values are stored in a cumulative distribution map. Distribution assumes valid values to be larger than 0 (not designed for signed measures), and normalized should the values not add up to 1. Probably not very efficient for large number of points, and for this purpose a different solution should be used. It is not recommended that this distribution is used yet since a few (potentially) problems could exist with it!!

Field Summary

Fields inherited from class IRPact_modellierung.distributions.BoundedUnivariateDistribution

lowerBound, upperBound

Fields inherited from class IRPact_modellierung.distributions.Distribution

name

Constructor Summary

Constructors

Constructor and Description
FiniteMassPointsDiscreteDistribution(java.lang.String name, java.util.Map<java.lang.Double,java.lang.Double> massPoints) Generates the cumulative distribution corresponding to the mass points if all mass points are valid (non-negative mass, since signed measures are outside of the scope of this distribution).

Method Summary

Modifier and Type	Method and Description
double	draw() Draws a value from the underlying distribution.
double	drawValue() Method that generates the values from the underlying distributions for the draw-method.
java.util.Map<java.lang.Double,java.lang.Double>	getCumulatedDistribution()
java.util.Map<java.lang.Double,java.lang.Double>	getMassPoints()

Methods inherited from class IRPact_modellierung.distributions.BoundedUnivariateDistribution

getLowerBound, getUpperBound

Methods inherited from class IRPact_modellierung.distributions.Distribution

getName, toString

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, wait, wait, wait

Constructor Detail

FiniteMassPointsDiscreteDistribution

public FiniteMassPointsDiscreteDistribution(java.lang.String name,
                                            java.util.Map<java.lang.Double,java.lang.Double> massPoints)
                                     throws java.lang.IllegalArgumentException

Generates the cumulative distribution corresponding to the mass points if all mass points are valid (non-negative mass, since signed measures are outside of the scope of this distribution). Will ignore points with 0 mass and normalize the massPoints provided

Parameters:

name - Name of the distribution (will be prefixed with FiniteMassPointsDiscreteDistribution_)

massPoints - Mass points as a map (x, mass(x)), must have non-negative mass

Throws:

java.lang.IllegalArgumentException - will be thrown when an illegal mass point (negative mass) is encountered

Method Detail

getMassPoints

public java.util.Map<java.lang.Double,java.lang.Double> getMassPoints()

getCumulatedDistribution

public java.util.Map<java.lang.Double,java.lang.Double> getCumulatedDistribution()

draw

public double draw()

Description copied from class: BoundedUnivariateDistribution

Draws a value from the underlying distribution. Will keep drawing values until it is provided with a legal value (that lies in [lowerBound, upperBound]).

Overrides:

draw in class BoundedUnivariateDistribution

Returns:

Legal value of the distribution (within [lowerBound, upperBound]).

drawValue

public double drawValue()

Description copied from class: BoundedUnivariateDistribution

Method that generates the values from the underlying distributions for the draw-method. Will be called to derive values of instances

Specified by:

drawValue in class BoundedUnivariateDistribution

Returns:

A value provided by the underlying distribution